CODEWITHSUNDEEP
pythonnumpy
Python Learner
Python Learner

Reputation: 437

Error in numpy argmax with sparse matrix

I'm using np.argmax function in one of my python programme for one of my data frame which is completely sparse matrix.my data frame contains 253 rows & 22 column. The head function return this sample data set

  var1                     var2     Var3  var4 ... .. var18  var19 ... var22
    0                        0       0     0   ......  0      1    ...  0
    0                        0       0     0   ......  0      0    ...  1
    0                        0       1     0   ......  0      0    ...  0
    0                        0       0     1   ......  0      0    ...  0
    0                        0       0     0   ......  1      0    ...  0

I'm using following code

y=np.argmax(train_y, axis=1)

While running this code I'm getting error message 

ValueError: Shape of passed values is (1, 253), indices imply (22, 253)

Can you expert please help me to solve this problem?

Thanks in advance

Upvotes: 0

Views: 814

Answers (1)

Clock Slave
Clock Slave

Reputation: 7977

The code works fine for me when I use - 

y=np.argmax(train_y.values, axis=1)

The np.argmax is a numpy function, passing in a dataframe could be causing this error. Instead convert it into a numpy nd array using the .values attribute of dataframes

Upvotes: 1

Related Questions